# Runprompt

> Use this skill when the user wants to run LLM prompts from .prompt files in the shell.

- Skill: `javimosch/runprompt` (Agent Skill)
- Install (CLI): `npx skillmds@latest add javimosch/runprompt`
- Raw SKILL.md: https://api.skillmd.com/api/skills/javimosch/runprompt/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: javimosch (https://skillmd.com/u/javimosch)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/javimosch/runprompt

---


# Runprompt Plugin

Run LLM [.prompt](https://google.github.io/dotprompt/) files from your shell with a single-file Python script.

## Commands

### Version
- `supercli runprompt self version` — Print runprompt version

### Run Prompt
- `supercli runprompt prompt run <prompt_file> [variables]` — Run a .prompt file with template variables

### Passthrough
- `supercli runprompt -- <args>` — Pass through to runprompt CLI

## Usage Examples

```bash
# Run a prompt file with variables
./runprompt hello.prompt '{"name": "World"}'

# Pass variables via stdin
echo '{"name": "World"}' | ./runprompt hello.prompt

# Use with other providers
export OPENAI_API_KEY="your-key"
./runprompt --model openai/gpt-4o hello.prompt

# Interactive chat mode
./runprompt --chat expert.prompt
```

## Installation

```bash
pip install runprompt
```

## Key Features

- Dotprompt template format support
- Multiple LLM providers (Anthropic, OpenAI, Google AI, OpenRouter)
- Structured JSON output with Picoschema
- Shell tool execution before prompts
- File attachments and glob patterns
- Response caching